Minimizing recovery time after a high availability event in a large-scale storage system

Patent metadata

Field	Value
Publication number	US-10929049-B2
Application number	US-201916380087-A
Country	US
Kind code	B2
Filing date	Apr 10, 2019
Priority date	Apr 10, 2019
Publication date	Feb 23, 2021
Grant date	Feb 23, 2021

Abstract

In one aspect, minimization of recovery time after a high availability (HA) event in a storage system is provided. An aspect includes creating, during an active replication session, mini replication session MOM objects from replication session-related configuration information of the active replication session, and creating mini volume objects for volumes of the active replication session, Upon determining an occurrence of the HA corresponding to the data module, an aspect includes reconfiguring the mini replication session MOM objects and the mini volume objects prior to activation of the data module and prior to enabling input/output (TO) servicing by the data module. An aspect further includes performing non-replication IO-related configuration for the data module and, upon completion of the non-replication IO-related configuration, enabling the IO servicing by the data module. Subsequent to enabling the IO servicing by the data module, an aspect also includes loading and reconstructing remaining replication session-related configuration.

First claim

We claim:

1. A method for minimizing recovery time after a high availability (HA) event in a storage system, the method comprising:
during an active replication session,
creating in persistent storage mini replication session MOM objects from replication session-related configuration information of the active replication session, the mini replication session MOM objects including a subset of parameters needed to configure a data path of a data module in the storage system; and
creating in the persistent storage mini volume objects for volumes of the active replication session, the mini volume objects including a subset of parameters of volume configuration information needed to configure the data module for the active replication session; and

upon determining an occurrence of a HA event corresponding to the data module, the HA event causing a failure of the data module,
prior to activation of the failed data module and prior to enabling input/output (IO) servicing by the failed data module, reconfiguring the failed data module based on the mini replication session MOM objects and the mini volume objects;
performing non-replication IO-related configuration for the failed data module and, upon completion of the non-replication IO-related configuration, enabling the IO servicing by the failed data module; and
subsequent to enabling the IO servicing by the failed data module, loading and reconstructing remaining replication session-related configuration other than the enabled IO servicing for the failed data module.

2. The method of claim 1 , further comprising:
determining the high availability event also corresponds to a management module of the storage system, the method further comprising:
upon restart of the management module, and prior to reconfiguring the mini replication session MOM objects and the mini volume objects, loading the mini replication session MOM objects and the mini volume objects at a beginning of a repository load from the persistent storage; and
loading non-replication IO-related configuration information.

3. The method of claim 1 , further comprising:
upon termination of the active replication session, removing the mini replication session MOM objects and the mini volume objects from volatile memory and the persistent storage.

4. The method of claim 1 , wherein the active replication session is implemented in one of synchronous replication mode and metro replication mode.

5. The method of claim 1 , wherein the mini replication session MOM objects include:
parameters of remote storage systems to data that is to be mirrored as part of the active replication session, the parameters including at least one of remote system identifiers (IDs) and pairing labels used to identify the active replication session in local storage systems and the remote storage systems.

6. The method of claim 1 , wherein the mini volume objects include at least one of:
volume pairing labels used by both local storage systems and remote storage systems to identify a volume within a storage array; and
volume external identifiers (IDs) as seen by initiators.

7. A system for minimizing recovery time after a high availability (HA) event in a storage system, the system comprising:
a memory comprising computer-executable instructions; and
a processor operable by a storage system, the processor executing the computer-executable instructions, the computer-executable instructions when executed by the processor cause the processor to perform operations comprising:
during an active replication session,
creating in persistent storage mini replication session MOM objects from replication session-related configuration information of the active replication session, the mini replication session MOM objects including a subset of parameters needed to configure a data path of a data module in the storage system; and
creating in the persistent storage mini volume objects for volumes of the active replication session, the mini volume objects including a subset of parameters of volume configuration information needed to configure the data module for the active replication session; and

upon determining an occurrence of a high availability HA event corresponding to the data module, the HA event causing a failure of the data module,
prior to activation of the failed data module and prior to enabling input/output (IO) servicing by the failed data module, reconfiguring the failed data module based on the mini replication session MOM objects and the mini volume objects;
performing non-replication IO-related configuration for the failed data module and, upon completion of the non-replication IO-related configuration, enabling the IO servicing by the failed data module; and

subsequent to enabling the IO servicing by the failed data module, loading and reconstructing remaining replication session-related configuration other than the enabled IO servicing for the failed data module.

8. The system of claim 7 , wherein the high availability event also corresponds to a management module of the storage system, the operations further comprising:
upon restart of the management module, and prior to reconfiguring the mini replication session MOM objects and the mini volume objects, loading the mini replication session MOM objects and the mini volume objects at a beginning of a repository load from the persistent storage; and
loading non-replication IO-related configuration information.

9. The system of claim 7 , wherein upon termination of the active replication session, the operations further comprise removing the mini replication session MOM objects and the mini volume objects from volatile memory and the persistent storage.

10. The system of claim 7 , wherein the active replication session is implemented in one of synchronous replication mode and metro replication mode.

11. The system of claim 7 , wherein the mini replication session MOM objects include:
parameters of remote storage systems to data that is to be mirrored as part of the active replication session, the parameters including at least one of remote system identifiers (IDs) and pairing labels used to identify the active replication session in local storage systems and the remote storage systems.

12. The system of claim 7 , wherein the mini volume objects include at least one of:
volume pairing labels used by both local storage systems and remote storage systems to identify a volume within a storage array; and
volume external identifiers (IDs) as seen by initiators.

13. A computer program product embodied on a non-transitory computer readable medium, the computer program product including instructions that, when executed by a computer, causes the computer to perform operations comprising:
during an active replication session at a storage system,
creating in persistent storage mini replication session MOM objects from replication session-related configuration information of the active replication session, the mini replication session MOM objects including a subset of parameters needed to configure a data path of a data module in the storage system; and
creating in the persistent storage mini volume objects for volumes of the active replication session, the mini volume objects including a subset of parameters of volume configuration information needed to configure the data module for the active replication session; and

upon determining an occurrence of a high availability (HA) event corresponding to the data module, the HA event causing a failure of the data module,
prior to activation of the failed data module and prior to enabling input/output (IO) servicing by the failed data module, reconfigurin

Assignees

• Emc Ip Holding Co Llc

Inventors

• Kronrod Svetlana
• Chen Xiangping
• Harduf Yuval

Classifications

• G06F3/0644

  Management of space entities, e.g. partitions, extents, pools · CPC title

• G06F3/067 Primary

  Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS] · CPC title

• G06F3/0635

  by changing the path, e.g. traffic rerouting, path reconfiguration · CPC title

• G06F3/0632

  by initialisation or re-initialisation of storage systems · CPC title

• G06F3/0611

  in relation to response time · CPC title